Body mass index as a predictor of hepatic steatosis in living liver donors

ORIGINAL ARTICLES

Body Mass Index as a Predictor of Hepatic Steatosis in Living Liver Donors Mary E. Rinella,* Estella Alonso,† Sambasiva Rao,‡ Peter Whitington,† Jonathan Fryer,§ Michael Abecassis,§ Riccardo Superina,§ Steven L. Flamm,* and Andres T. Blei* Evaluation of the living donor for liver transplantation is a complex process involving such invasive studies as liver biopsy and angiography. It is important to establish the likelihood and extent of hepatic steatosis in living donors by clinical, imaging, and biochemical parameters to avoid performing a liver biopsy, if possible. In this study, the predictive value of body mass index (BMI), liver chemistry tests, and imaging studies was compared with liver histological examination in 33 potential living donors. Patients were grouped and compared based on their BMI (<25, 25 to 28, >28). No patient with a BMI less than 25 had hepatic steatosis. Of patients with a BMI of 25 to 28, steatosis was found on biopsy in 3 of 9 patients. Thirteen of 17 patients (76%) with a BMI greater than 28 had hepatic steatosis on liver biopsy. There was a significant correlation between BMI and overall grade of steatosis (R ⴝ 0.49). All subjects with steatosis detected on magnetic resonance imaging (MRI) or computed tomography (CT) had steatosis on biopsy, and all but 2 such patients had greater than 10% steatosis on biopsy. Conversely, 30% of patients in the MRI group and 24% of patients in the CT group failed to show hepatic steatosis when it was present on biopsy. Thus, it appears that liver biopsy could be avoided in subjects with a normal BMI and absence of risk factors. Individuals with a high BMI should undergo liver biopsy because biochemical and imaging data are currently inadequate to determine the extent of steatosis. Future studies should aim at improving the sensitivity of imaging techniques in the diagnosis of steatosis. (Liver Transpl 2001;7:409-414.)

E

valuation of the living liver donor can be an invasive process with inherent risk. Evaluation typically involves phlebotomy; imaging of arterial, venous, and biliary anatomy; and liver biopsy. Liver biopsy is performed to evaluate abnormal liver chemistry test results and screen for the presence of underlying hepatic pathological states. Although liver biopsy is generally safe (1% risk for serious complications), as many as 5% of patients may require hospitalization after the procedure.1,2 Therefore, it is of interest to delineate the utility of a liver biopsy in the workup of a living donor. Hepatic macrosteatosis is a common finding in cadaveric donor liver biopsy specimens, and when excessive, it places the recipient at risk for primary graft nonfunction.3-6 The effect of hepatic steatosis in living liver donation has not been well delineated. Marcos et

al7 proposed a relationship that every 1% of hepatic steatosis is assumed to decrease the functional graft mass by 1%. If the graft is large enough to account for the degree of steatosis, this group proceeds with the donor surgery. More recently, Marcos et al8 published a series in which no impairment in function was found in either the living donor or recipient using grafts containing less than 30% steatosis. Other groups only accept living donor grafts with less than 20% macrosteatosis.9 Preoperative determination of the extent of hepatic steatosis is important. Liver biopsy is currently the gold standard, although noninvasive means are being sought. Previous studies have shown a positive correlation between hepatic steatosis and body mass index (BMI).10,11 Other methods, including body topography and alanine aminotransferase (ALT) and serum lipid levels, correlated with hepatic steatosis, although weakly.10,12 Specifically, one study found an elevated waist to hip ratio to be a better predictor of fatty infiltration than BMI.13 However, none of these methods correlated unequivocally with hepatic steatosis. It is possible that biochemical parameters in conjunction with hepatic imaging could accurately predict the presence and extent of steatosis and potentially avoid liver biopsy. In this study, we compared the predictive value of BMI, liver chemistry test results, and imaging studies with liver histological examination in potential living liver donors to determine the presence and severity of hepatic steatosis. From the Departments of *Medicine, §Surgery, †Pediatrics, and ‡Pathology, Northwestern Memorial Hospital and Northwestern University, Chicago, IL. Address reprint requests to Andres T. Blei, MD, Northwestern Medical Faculty Foundation, 675 N St Clair, Galter 15-250, Chicago, IL 60611. Telephone: 312-695-9286; FAX: 312-695-5998; E-mail: [email protected] Copyright © 2001 by the American Association for the Study of Liver Diseases 1527-6465/01/0705-0002$35.00/0 doi:10.1053/jlts.2001.23787

Liver Transplantation, Vol 7, No 5 (May), 2001: pp 409-414

409

410

Rinella et al

Subjects and Methods Subjects Between September 1997 and October 2000, a total of 53 subjects underwent evaluation for living liver donation at Northwestern Memorial Hospital (Chicago, IL). The evaluation algorithm includes 4 phases: information session and blood type (phase I), history and physical examination by an unbiased physician and blood work (phase II), liver imaging and biopsy (phase III), and meeting with the transplant surgeon (phase IV). The charts of 33 potential donors who progressed to phase III of the evaluation protocol were retrospectively reviewed. All 33 patients (19 men, 14 women) underwent both hepatic imaging and liver biopsy. Potential donors underwent a full medical history and physical evaluation and none had known comorbidities, specifically diabetes. No subject was on chronic medications, and none consumed alcohol in excess. Twenty-one of these 33 subjects included in the analysis were believed suitable for donation based on blood type, anatomy, absence of significant hepatic steatosis, and willingness to donate. Many patients were excluded before they completed phase III of evaluation because of abnormal anatomy and are thus not included in the analysis. Of those selected for donation, 20 subjects donated to a child and 1 subject donated the right lobe to an adult.

Medical Evaluation All subjects had serum liver chemistry and complete blood count determinations. Most had serum lipid levels measured. BMI was calculated for all patients. Donor age ranged from 18 to 47 years. Parameters of body weight and height, laboratory values, and hepatic imaging were compared with liver biopsy results in the same individual. Potential donors underwent either computed tomography (CT) and angiography or magnetic resonance (MR) imaging (MRI), MR angiography, and/or MR venography to evaluate the presence of hepatic fat, delineate the anatomy, and assess liver volume. All MR studies were performed using a 1.5 Tesla Siemens Sonata MRI scanner (Siemens Corp, New York, NY). The presence of hepatic steatosis by MRI was assessed by chemical shift imaging with T1 in-phase and oppose-phase images. Quantification of hepatic steatosis was not calculated in this study. The role of MR studies in the complete evaluation of the living donor is reported elsewhere.14 If the potential donor had either normal or near-normal liver chemistry test results and no significant anatomic abnormalities that precluded donation, liver biopsy was performed. A single pass was made into the right lobe using a Bard Monopty needle (C.R. Bard Inc, Billerca, MA) after ultrasound localization. The average length of biopsy specimens when measured after fixation in formalin was 1.5 cm. The same pathologist (S.R.) reviewed hematoxylin and eosin–stained sections of liver biopsy specimens on 2 occasions for fat quantitation in a blinded fashion. A second pathologist evaluated discrepancies noted by the first. The final reading was that of the second pathologist in cases of discrep-

ancy with the first pathologist. Three of the 33 biopsy specimens had intraobserver variability and were thus finalized by the second pathologist. Neither pathologist had information about the clinical characteristics of the patients nor the previous biopsy reading. The steatosis grading scale previously described by D’Alessandro et al15 was used as a base. This gradation reflects an overall assessment of fat accumulation in the specimen. In this study, the grading system was modified slightly, underscoring the upper limit of steatosis acceptable for transplantation in living donation at our institution. Although an acceptable limit of hepatic macrosteatosis in living liver donation has not been firmly established, in our institution only living donor livers with less than 10% hepatic macrosteatosis are used. For this reason, the grading was modified as grade 0 (no steatosis), grade 1a (ⱕ10%), grade 1b (11% to 30%), grade 2 (31% to 60%), and grade 3 (⬎60%) macrosteatosis on the biopsy specimen. Comparison of groups within the cohort was made by means of 1-way ANOVA. Differences were evaluated using Tukey’s test. Results are reported as mean ⫾ SD.

Results Patient Characteristics The characteristics of the 33 patients who reached phase III of the evaluation process and underwent liver biopsy are listed in Table 1. Subjects were divided into 3 groups based on BMI: BMI less than 25 (normal), BMI of 25 to 28 (overweight), and BMI greater than 28 (obese). The mean age of the patients was similar between groups, as was height. By definition, weight was significantly greater in groups with the 2 greater BMIs (P ⬍ .001). The percentage of patients with a BMI greater than 25 in those who reached phase III of the evaluation was similar to the percentage of overweight patients in the initial 53 subjects presenting for evaluation (75% and 78%, respectively). Biochemical Parameters All patients had hepatic biochemical tests performed during evaluation. Of patients with a BMI less than 25, BMI of 25 to 28, and BMI greater than 28, ALT levels exceeded the upper limit of normal in 0 of 7 patients, 1 of 9 patients (11%), and 3 of 17 patients (18%), respectively. When compared, no statistically significant difference was found. Two of 15 patients without steatosis on their biopsy specimen had ALT levels greater than normal (1 patient had an ALT level of 49 IU/mL, 1 unit greater than the upper limit of normal). Three of 5 patients with grades 2 and 3 steatosis on biopsy had normal ALT levels (Table 2). Other liver biochemical parameters, including aspartate aminotransferase (AST), alkaline phosphatase, and ␥-glutamyltransferase

411

BMI and Hepatic Steatosis in the Living Donor

Table 1. Patient Characteristics Group

Age (yr)

Height (cm)

Weight (kg)

ALT (IU/mL)

AST (IU/mL)

AP (IU/mL)

GGT (IU/mL)

33.1 4.5

173.1 6.8

65.2 6.9

18.3 6.9

27.7 19.7

81.2 40.9

24.9 25.2

28.7 10.1

174.2 10.7

81.5* 10.1

24.3 19.4

23.4 7.9

74.6 21.9

19.0 11.0

35.3 7.1

171.4 10.1

91.7* 12.9

31.5 15.1

26.3 6.7

71.9 20.3

27.0 14.9

BMI ⬍ 25 (3 M/4 F) Mean SD BMI 25-28 (6 M/3 F) Mean SD BMI ⬎ 28 (10 M/7 F) Mean SD

NOTE. Normal laboratory values: AST (6 to 40 IU/mL), ALT (8 to 48 IU/mL), alkaline phosphatase (30 to 115 IU/mL), and ␥-glutamyltransferase (10 to 32 IU/mL). Abbreviations: AP, alkaline phosphatase; GGT, ␥-glutamyltransferase. * Differences between groups, P ⬍ .001.

No patient with a BMI less than 25 had hepatic steatosis. Three of 9 patients with a BMI of 25 to 28 had steatosis on biopsy: One patient had grade 1a (1% to 10% fat), 1 patient had grade 1b (11% to 30% fat), and

1 patient had grade 2 (31% to 60% fat). The patient with grade 2 steatosis had a BMI of 28 (upper limit for this group). Two of 3 patients had an imaging study result negative for steatosis and thus needed biopsy to elucidate the presence of fat (1 patient, grade 1a steatosis; 1 patient, grade 2 steatosis). Patients with a BMI greater than 28 were most likely to have fat on biopsy; 13 of 17 patients (76%) had steatosis identified on liver biopsy. Eight of the 13 patients with documented hepatic steatosis on biopsy had grade 1 steatosis. Quantitative fat breakdown showed 6 patients had less than 10% fat (grade 1a), and 2 patients, 30% (grade 1b) fat on biopsy. This group of patients with a BMI greater than 28 were also the most likely to have grades 2 and 3 steatosis (⬎30% fat; Fig. 1) There was a significant correlation between BMI and overall grade of steatosis (R ⫽ 0.49; P ⬍ .01), and this was slightly more pronounced in women (R ⫽ 0.52). Differences in the degree of steatosis were significant on ANOVA (F ⫽ 3.55; P ⬍ .01), and Tukey’s test showed differences between the groups with a BMI less than 25 and BMI greater than 28 (P ⬍ .01).

Table 2. ALT and Steatosis Grade

Table 3. Hepatic Imaging and Steatosis

levels, were also poorly sensitive and specific in predicting the presence or absence of fat in the liver. Imaging In this study, 17 of 33 patients underwent MRI and 18 of 33 patients underwent CT (3 patients underwent both, 1 patient did not undergo imaging) to evaluate the presence of fat (Table 3), liver volume, and biliary, venous, and arterial anatomy. When imaging studies reported steatosis, all patients had steatosis on biopsy. When imaging studies reported no fat, 3 of 10 MRI studies (30%) and 4 of 17 examinations by CT (24%) had hepatic steatosis on biopsy. In such subjects, the 3 patients in the MRI group had grade 1a steatosis; 3 of the 4 patients in the CT group had grade 1a steatosis, and the other patient had grade 2 steatosis. Effect of BMI on the Presence of Steatosis

Steatosis Grade

Normal ALT Abnormal ALT

Hepatic Steatosis

0

1a

1b

2-3

15 2

6 1

3 0

3 2

NOTE. Steatosis grading: grade 0, no macrosteatosis; grade 1a, less than 10% macrosteatosis; grade 1b, 11% to 30% macrosteatosis; grade 2, 31% to 60% macrosteatosis; grade 3, greater than 60% macrosteatosis.

MRI No fat Fat CT No fat Fat

Steatosis Grade

Present

Absent

0

1a

1b

2

3

3 7

7 0

7 0

3 2

0 2

0 2

0 1

4 1

13 0

13 0

3 0

0 0

1 0

0 1

412

Rinella et al

Figure 1. All subjects with a BMI less than 25 had grade 0 steatosis. As BMI increased, more subjects had a greater degree of steatosis. Steatosis grading: grade 0, no macrosteatosis; grade 1a, less than 10% macrosteatosis; grade 1b, 11% to 30% macrosteatosis; grade 2, 31% to 60% macrosteatosis; and grade 3, greater than 60% macrosteatosis.

Recipient and Donor Outcome Twenty-one subjects underwent partial hepatectomy for donation. Five of the 21 subjects had grade 1a steatosis, and the remainder had none. One of the 5 donors with grade 1a steatosis underwent resection; however, the child died before the graft was transplanted. Of the remaining 4 recipients of grafts with grade 1 steatosis, 3 recipients are alive and well and 1 child died of postoperative hemorrhage. In the 16 recipients of grafts with grade 0 steatosis, there were no cases of graft primary nonfunction or primary dysfunction.

Discussion A thorough medical evaluation of the potential living donor is important to ensure both donor and recipient safety. Hepatic steatosis has become a focal point in the field of living donor transplantation because of its potential ramifications on graft function and recipient outcome. Liver biopsy is the gold standard for the detection of hepatic steatosis in the evaluation of the living donor. Although liver biopsy is considered a relatively safe procedure, up to 5% will require hospitalization after the procedure, 1% will have significant bleeding, and 1 in 10,000 will have a fatal outcome.1,2 However small these risks may be, it would be optimal to avoid performing a biopsy because increasing numbers of donors will be evaluated for living liver transplantation in the future. Because of the imperfect nature of imaging,

biochemical, and other ancillary parameters in the detection of steatosis, a policy could be adopted in which a liver biopsy is always indicated. However, one should weigh the potential risks of biopsy with the unlikely possibility of finding unexpected significant liver pathology in the operating room. Accurate detection and quantification of hepatic fat are essential in determining the potential donor’s eligibility. In cadaveric transplants, the presence of significant macrosteatosis (⬎60%) has been associated with primary nonfunction of the graft.15,16 Others have shown diminished patient survival posttransplantation in recipients of grafts with less marked steatosis.4 The degree of steatosis acceptable in living donor livers is controversial. Marcos et al8 found no impairment of either liver function or graft regeneration in recipients of grafts with less than 30% steatosis on biopsy. However, donors with significant steatosis may not tolerate surgery as well as those with nonsteatotic livers and may have impaired hepatic regeneration. This concept has been shown in an animal model of partial hepatectomy17 in which obese Zucker rats with steatotic livers had significantly delayed regeneration after 70% hepatectomy, which in turn led to diminished animal survival. In addition, a retrospective study of humans after major hepatic resection found moderate to severe hepatic steatosis was associated with increased postoperative morbidity, mortality, transfusion requirement, and surgical time.18 Hepatic steatosis is a common feature in cadaveric liver donors.19 Our data support the notion that hepatic steatosis is common among living donors, in great part related to the high prevalence of obesity in the general American population. Not only are patients with a BMI greater than 25 more likely to have hepatic steatosis, they are at risk for showing fibrosis or even indolent cirrhosis on liver biopsy.20 Seventy-five percent of our donor population was overweight, defined as a BMI greater than 25. This illustrates an important problem in the donor population and it approaches the crude prevalence of individuals with a BMI greater than 25 in the United States.21 Such ancillary parameters as biochemical profiles, anthropometric measurements, and hepatic imaging could be useful in elucidating the presence of fat on liver biopsy.12,22 However, all parameters are imperfect. Of biochemical parameters, triglyceride and cholesterol levels have shown the most promise. Several studies have shown an association between triglyceride levels and hepatic steatosis.23,24 Not all our potential donors had serum lipid levels assessed in the fasting state; therefore, we did not include serum lipid levels in our anal-

BMI and Hepatic Steatosis in the Living Donor

ysis. We determined whether there was a correlation between ALT, AST, alkaline phosphatase, and ␥-glutamyltransferase levels and BMI. Subjects with a greater BMI appeared to have a trend toward a greater ALT level, as previously noted.25 However, there was no statistically significant association between ALT level and the presence of steatosis on biopsy. From our data, it appears that liver biochemical tests are not good predictors of hepatic steatosis. The ratio of ALT to AST levels has been shown to correlate with the presence of fat on liver biopsy.26 In our study, no such correlation was found. Our data suggest that BMI is a more reliable predictor of the presence of steatosis. There was a positive correlation between increasing BMI and steatosis grade on biopsy. Other studies have found a greater prevalence of steatosis in patients with increased BMI.10,20,24,27 One intriguing study examined the effect of truncal fat distribution (reflected by the waistto-hip ratio) and showed a stronger correlation (R ⫽ .44) with steatosis on biopsy than BMI (R ⫽ .04).13 Multiple imaging modalities have been evaluated for their ability to detect hepatic fat. Some of the investigated imaging tests include ultrasound, CT, and MR studies.28-30 In this study, patients underwent either CT or MRI as part of their evaluation. In our experience, when imaging detected the presence of fat, it was present on biopsy in all cases. However, in 40% (MRI) and 24% (CT) of cases, imaging failed to show steatosis when it was present. Although imaging is a useful adjunct to liver biopsy in the evaluation of the living liver donor, at this time it does not replace it. Seventy-eight percent of patients with fat detected on imaging had enough steatosis on biopsy to preclude donation at our institution (⬎10% fat). However, 2 of the patients with fat by imaging had minimal fat and thus were candidates for donation and would have not qualified as a potential donor had a biopsy not been performed. Individuals with a BMI of 25 to 28 and imaging that shows hepatic steatosis should undergo liver biopsy because they are less likely than their more overweight counterparts to have a high degree of steatosis on liver biopsy. Subjects with a BMI greater than 28 tend to have a greater degree of hepatic steatosis; however, imaging did not discriminate the extent of fat accumulation in such individuals. Future studies examining more precise fat quantitation by MR studies or CT are needed to more thoroughly address this issue. No subject with a BMI less than 25 in this study had fat detected on imaging or steatosis on biopsy. It appears from our results that liver biopsy may be avoided in such individuals. Potential donors with fat detected on MR studies and CT may still be evaluated as possible

413

candidates because 20% of our subjects in this category did not have enough fat to disqualify them from donation. If time permits, potential donors can be given the opportunity to lose weight and be reevaluated because weight loss has been shown to decrease hepatic steatosis.31,32 The threshold of hepatic steatosis acceptable for living liver donation has not been conclusively established. At our center, we use only livers with less than 10% steatosis on biopsy for living donor transplantation. Although this approach may seem conservative, we believe it both minimizes surgical risk for the donor and reduces the possibility of a poor outcome related to a steatotic graft in the recipient. Other groups exclude donors when greater than 30% steatosis is detected, and some9 use a cutoff value of 20% in living donors. Until more centers publish their experience with living donor grafts with varying degrees of steatosis, the acceptable limit on steatosis in the living donor graft will remain uncertain. In conclusion, it appears that liver biopsy can be avoided in subjects with a normal BMI and absence of other comorbid conditions, such as diabetes or hyperlipidemia. The majority of candidates undergoing evaluation for living related donation with a high BMI should undergo a liver biopsy because biochemical and imaging data are not reliable enough to accurately diagnose the degree of steatosis.

Acknowledgment The authors thank Lori Clark for her indispensable involvement with the donors and help accruing and organizing the data.

References 1. Van Ness MM, Diehl AM. Is liver biopsy useful in the evaluation of patients with chronically elevated liver enzymes? Ann Intern Med 1989;111:473-478. 2. Goddard CJ, Warnes TW. Raised liver enzymes in asymptomatic patients: Investigation and outcome. Dig Dis 1992;10:218226. 3. Urena MA, Ruiz-Delgado FC, Gonzalez EM, Segurola CL, Romero CJ, Garcia IG, et al. Assessing risk of the use of livers with macro and microsteatosis in a liver transplant program. Transplant Proc 1998;30:3288-3291. 4. Marsman WA, Wiesner RH, Rodriguez L, Batts KP, Porayko MK, Hay JE, et al. Use of fatty donor liver is associated with diminished early patient and graft survival. Transplantation 1996;62:1246-1251. 5. D’Alessandro AM, Kalayoglu M, Sollinger HW, Hoffmann RM, Reed A, Knechtle SJ, et al. The predictive value of donor liver biopsies on the development of primary nonfunction after ortho-

414

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

Rinella et al

topic liver transplantation. Transplant Proc 1991;23:15361537. Ploeg RJ, D’Alessandro AM, Knechtle SJ, Stegall MD, Pirsch JD, Hoffmann RM, et al. Risk factors for primary dysfunction after liver transplantation—A multivariate analysis. Transplantation 1993;55:807-813. Marcos A, Ham JM, Fisher RA, Olzinski AT, Posner MP. Single-center analysis of the first 40 adult-to-adult living donor liver transplants using the right lobe. Liver Transpl 2000;6:296-301. Marcos A, Fisher RA, Ham JM, Shiffman ML, Sanyal AJ, Luketic VA, et al. Liver regeneration and function in donor and recipient after right lobe adult to adult living donor liver transplantation. Transplantation 2000;69:1375-1379. Fan ST, Lo CM, Liu CL, Yong BH, Chan JK, Ng IO. Safety of donors in live donor liver transplantation using right lobe grafts. Arch Surg 2000;135:336-340. Guzzaloni G, Grugni G, Minocci A, Moro D, Morabito F. Liver steatosis in juvenile obesity: Correlations with lipid profile, hepatic biochemical parameters and glycemic and insulinemic responses to an oral glucose tolerance test. Int J Obes Relat Metab Disord 2000;24:772-776. Monsour HP Jr, Wood RP, Ozaki C, Katz S, Clark J, Dyer C, et al. Utility of preoperative liver biopsy in live-related donor patients for liver transplantation. Transplant Proc 1994;26:138139. Bellentani S, Saccoccio G, Masutti F, Croce LS, Brandi G, Sasso F, et al. Prevalence of and risk factors for hepatic steatosis in Northern Italy. Ann Intern Med 2000;132:112-117. Kral JG, Schaffner F, Pierson RN Jr, Wang J. Body fat topography as an independent predictor of fatty liver. Metabolism 1993; 42:548-551. Flamm S, Alonso E, Blei AT, Whitington P, Stuart F, Superina R, et al. MR imaging of the liver: A cost-effective non-invasive single approach to living donor evaluation. Liver Transpl 2000; 6:C13. D’Alessandro AM, Kalayoglu M, Sollinger HW, Hoffmann RM, Reed A, Knechtle SJ, et al. The predictive value of donor liver biopsies for the development of primary nonfunction after orthotopic liver transplantation. Transplantation 1991;51: 157-163. Todo S, Demetris AJ, Makowka L, Teperman L, Podesta L, Shaver T, et al. Primary nonfunction of hepatic allografts with preexisting fatty infiltration. Transplantation 1989;47:903-905. Selzner M, Clavien PA. Failure of regeneration of the steatotic rat liver: Disruption at two different levels in the regeneration pathway. Hepatology 2000;31:35-42. Behrns KE, Tsiotos GG, DeSouza NF, Krishna MK, Ludwig J, Nagorney DM. Hepatic steatosis as a potential risk factor for major hepatic resection. J Gastrointest Surg 1998;2:292-298.

19. Urena G. Hepatic steatosis in liver transplant donors: Common feature of donor population? World J Surg 1998;22:837-844. 20. Ratziu V, Giral P, Charlotte F, Bruckert E, Thibault V, Theodorou I, et al. Liver fibrosis in overweight patients. Gastroenterology 2000;118:1117-1123. 21. Flegal KM, Carroll MD, Kuczmarski RJ, Johnson CL. Overweight and obesity in the United States: Prevalence and trends, 1960-1994. Int J Obes Relat Metab Disord 1998;22:39-47. 22. Liu TY, Lu SN, Su WP, Chang WY, Wang LY, Hsieh MY, et al. Prediction of fatty liver from serum triglyceride levels and body weight indexes. Kao Hsiung I Hsueh Ko Hsueh Tsa Chih 1990; 6:289-294. 23. Banerji MA, Buckley MC, Chaiken RL, Gordon D, Lebovitz HE, Kral JG. Liver fat, serum triglycerides and visceral adipose tissue in insulin-sensitive and insulin-resistant black men with NIDDM. Int J Obes Relat Metab Disord 1995;19:846-850. 24. Luyckx FH, Desaive C, Thiry A, Dewe W, Scheen AJ, Gielen JE, et al. Liver abnormalities in severely obese subjects: Effect of drastic weight loss after gastroplasty. Int J Obes Relat Metab Disord 1998;22:222-226. 25. Piton A, Poynard T, Imbert-Bismut F, Khalil L, Delattre J, Pelissier E, et al. Factors associated with serum alanine transaminase activity in healthy subjects: Consequences for the definition of normal values, for selection of blood donors, and for patients with chronic hepatitis C. MULTIVIRC Group. Hepatology 1998;27:1213-1219. 26. Nanji AA, French SW, Freeman JB. Serum alanine aminotransferase to aspartate aminotransferase ratio and degree of fatty liver in morbidly obese patients. Enzyme 1986;36:266-269. 27. Angulo P, Keach JC, Batts KP, Lindor KD. Independent predictors of liver fibrosis in patients with nonalcoholic steatohepatitis. Hepatology 1999;30:1356-1362. 28. Ricci C, Longo R, Gioulis E, Bosco M, Pollesello P, Masutti F, et al. Noninvasive in vivo quantitative assessment of fat content in human liver. J Hepatol 1997;27:108-113. 29. Levenson H, Greensite F, Hoefs J, Friloux L, Applegate G, Silva E, et al. Fatty infiltration of the liver: Quantification with phasecontrast MR imaging at 1.5 T vs biopsy. AJR Am J Roentgenol 1991;156:307-312. 30. Pamilo M, Sotaniemi EA, Suramo I, Lahde S, Arranto AJ. Evaluation of liver steatotic and fibrous content by computerized tomography and ultrasound. Scand J Gastroenterol 1983;18: 743-747. 31. Andersen T, Gluud C, Franzmann MB, Christoffersen P. Hepatic effects of dietary weight loss in morbidly obese subjects. J Hepatol 1991;12:224-229. 32. Ueno T, Sugawara H, Sujaku K, Hashimoto O, Tsuji R, Tamaki S, et al. Therapeutic effects of restricted diet and exercise in obese patients with fatty liver. J Hepatol 1997;27:103-107.